Caryologia. International Journal of Cytology, Cytosystematics and Cytogenetics 73(4): 27-38, 2020 Firenze University Press www.fupress.com/caryologia ISSN 0008-7114 (print) | ISSN 2165-5391 (online) | DOI: 10.13128/caryologia-1133 Caryologia International Journal of Cytology, Cytosystematics and Cytogenetics Citation: E. Palchetti, M. Gori, S. Biri- colti, A. Masoni, L. Bini, C. Tani, S. Falsini, E. Corti, A. Papini (2020) Possible hybrid speciation for two Malagasy species of Piper L. (Piperaceae). Cary- ologia 73(4): 27-38. doi: 10.13128/caryo- logia-1133 Received: November 12, 2020 Accepted: November 15, 2020 Published: May 19, 2021 Copyright: © 2020 E. Palchetti, M. Gori, S. Biricolti, A. Masoni, L. Bini, C. Tani, S. Falsini, E. Corti, A. Papini. This is an open access, peer-reviewed article published by Firenze University Press (http://www.fupress.com/caryologia) and distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distri- bution, and reproduction in any medi- um, provided the original author and source are credited. Data Availability Statement: All rel- evant data are within the paper and its Supporting Information files. Competing Interests: The Author(s) declare(s) no conflict of interest. Author Contributions: Conceptualiza- tion, Palchetti, Biricolti and Gori; meth- odology, Gori, Papini; software, Papini; validation, Palchetti, Gori, Papini, Biri- colti, Calamai; formal analysis, Papini, Gori, Biricolti, Bini; investigation, Gori, Bini, Falsini, Corti, Calamai; resources, Palchetti; data curation, Gori; writing— original draft preparation, all authors; project administration, Palchetti; fund- ing acquisition, Palchetti. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the Company TozziGreen, La Tour - 26 Etage Rue Ravoninahitriniarivo – Ankorondrano, Antananarivo 101 Mad- agascar. Possible hybrid speciation for two Malagasy species of Piper L. (Piperaceae) Enrico Palchetti1, Massimo Gori1,4, Stefano Biricolti1,*, Alberto Masoni1, Lorenzo Bini4, Corrado Tani2, Sara Falsini2, Emilio Corti2, Alessio Papini2,3 1 Department of Agriculture, Food, Environment and Forestry (DAGRI) University of Florence, Piazzale delle Cascine, 18, 50144 Florence (FI), Italy; E-mail: enrico.palchetti@ unifi.it 2 Department of Biology, University of Florence, Via Micheli, 3, Firenze, Italy 3 CSET Tropical Herbarium University of Florence, Italy; E-mail: alpapini@unifi.it 4 Interdepartmental Service Centre for Agricultural, Chemical and Industrial Biotechnolo- gies (CIBIACI), University of Florence, Via Romana, 21, 50125 Florence, Italy: E-mail: massimo.gori@unifi.it *Corresponding author. E-mail: stefano.biricolti@unifi.it Abstract. Two new species of genus Piper L. from Madagascar: Piper malgassicum Pap- ini, Palchetti, M. Gori & Rota Nodari and Piper tsarasotrae Papini, Palchetti, M. Gori & Rota Nodari, were analyzed to investigate their phylogenetic position and evolution- ary history. Both plastidial and nuclear markers were used for sequencing. The plas- tidial markers (ndhF and trnL intron) showed a close relationship between the two species with respect to the other species of Piper. Both species appeared phylogeneti- cally related to the African P. guineense and the Malagasyan/Mascarenhas endemic P. borbonense. The nuclear marker (G3pdh) amplification produced two separate sets of sequences: “long” sequences and “short” sequences, characterized by some long dele- tions. Analyzing together the nuclear sequences, we observed that the “long” sequence of P. tsarasotrae had a stricter relationship to the African accessions of P. guineense, while the accession of P. malgassicum was more strictly related to P. borbonense. On the contrary both “short” sequences of P. malgassicum and P. tsaratsotrae resulted phylo- genetically related to Asian accessions and more distantly related to the formerly cited species. This unexpected result was tentatively explained with a more ancient hybrid- ization event between an ancestor of P. malgassicum and P. tsarasotrae (and possibly P. borbonense) and an Asian species of Piper. The Asian contribution would have pro- duced the ancestors carrying the “short” sequences. A more recent hybridization event would have led to the separation of P. malgassicum from P. tsarasotrae with an African pollen-derived genome contribution from P. guineense or, more probably, an ancestor thereof, to an ancestor of P. tsarasotrae. The chromosome numbers of P. tsarasotrae (2n = about 38) and P. malgassicum (2n = about 46), were more similar to the Asian spe- cies than to the American species. Unfortunately, no chromosome number of the Afri- can species P. guineense is currently available, to compare the chromosomal numbers. Keywords: Piper malgassicum, Piper tsarasotrae, Piperaceae, chromosomes, hybridiza- tion, DNA sequences, G3pdh, trnL, ndhF, Malagasy biodiversity. 28 Enrico Palchetti et al. 1. INTRODUCTION Genus Piper L. (Piperaceae) is one of the largest gen- era of Angiosperms, with more than 2000 species (Qui- jano-Abril et al. 2006) and were considered belonging to a basal group of angiosperms, the so called “paleoherbs” (Loconte et al. 1991). Piper is a pantropical genus developing highly vari- able growth forms (Isnard et al. 2012), with the highest biodiversity in the American continent with a number of species ranging from 500 (Burger 1972; Tebbs 1993), to 1100 (Jaramillo and Manos 2001), later increased to more than 1800 (Ulloa Ulloa et al. 2017), many of them with a small distribution area (Quijano-Abril et al. 2006). The separation of species is often tricky, due to the small size of the floral parts and hence the number of synonyms may be high (Suwanphakdee et al. 2016), while other species tend to get naturalized (Smith et al. 2008). While only two species are known as native of the African continent, P. guineense Thonn. and P. capense L. f., more species are known of Madagascar, even if some of them are known only for a single or few herbarium samples. The currently recognized species in Madagascar are P. heimii C. DC, P. pachyphyllum Baker and possibly P. borbonense (Miq.) C. DC., described for the Bourbon island, nowadays La Reunion (Weil et al. 2017), belonging to the Mascar- enhas Islands. However, its presence in Madagascar was affirmed by De Candolle (1869; 1923). The fact that P. bor- bonense is cultivated makes more complex to understand its real distribution area (Palchetti et al. 2018). Piper malgassicum Papini, Palchetti, M. Gori, Rota Nodari and Piper tsarasotrae Papini, Palchetti, M. Gori, Rota Nodari, were recently described as new Malagasy species (Palchetti et al. 2018) and are of economic inter- est, since their dried fruits are often mixed with P. bor- bonense to produce the typical Malagasy spice called in local language “voatsiperifery” pepper. The aim of the investigation was to understand how the malagasyan species might have been originated and their relationships with the African and the Asian spe- cies. This knowledge will help to understand how the Malagasyan species used as spices may be related to P. nigrum with possible future biotechnological implica- tions. The chosen method for anwering the research goal was the analysis of DNA sequences both of nuclear and plastidial origin and the chromosome numbers of P. malgassicum and P. tsarasotrae. 2. MATERIALS AND METHODS A first round of sample collection within the inter- nal area of Madagascar was conducted in 2016 and the samples have been submitted to analyses. The results have been reported in Palchetti et al. (2018) but, in order to get a deeper knowledge about the genetic asset of the two species and to confirm the obtained results e sec- ond round of sample collection has been carried on in 2019. 4 new plants were collected in two different are- as of the Ambositra region in Madagascar. The first 2 plants, belonging to the P. malgassicum type, were col- lected in the tropical rainy forest of Vohiday and the second 2 plants, belonging to the P. tsarasotrae type, in the semi-dry area of the Tsaratsotra village. These plants were compared with the samples of P. tsarasotrae and P. malgassicum which have been used for a previous inves- tigation that included the description of the species (Pal- chetti et al. 2018). Samples were conserved either in eth- anol 96% either as herbarium sample by the ET (Tropi- cal Herbarium of Florence, CSET, https://www.bio.unifi. it). Some seeds were also germinated in Florence for karyotyping. The DNA used for this work was extract- ed from tissue conserved in ethanol 96% (Murray et al. 1996; Bressan et al. 2014). DNA was extracted from 40 mg of the ethanol pre- served leaves after drying under vacuum. The starting material was inserted in 2 ml tube, together with tung- sten carbide beads, frozen in liquid nitrogen and finely ground in a tissue homogenizer (Tissue Lyser ®, Qiagen). DNA was extracted using Invisorb Spin Plant Mini kit (Stratec molecular®) according to the manifacture’s guidelines. Amplification of the trnL (UAA) intron (trnL) and the low copy nuclear gene glyceraldehyde 3-phosphate dehydrogenase (G3pdh) followed respectively the proto- cols by Taberlet et al. (1991) and Strand et al. (1997). Two new primer pairs were designed using the chloroplast genome sequence of P. kadsura (GenBank®: KT223569.1) as template to cover the entire NADH dehydrogenase F (ndhF) plastid gene: ndhF-F3_for- ward 5’-AGGTTCTTATCGAGCCGCTT-3’ and ndhF- F3_reverse 5’-GTAAGAAGAAATGCGCCCCC-3’ and ndhF-F10_forward 5’-CTTCGCCGTATGTGGGCTTT-3’ and ndhF-F10_reverse 5’-TCGACCAAAAGCAAGCAA- GAG-3’. The amplicons have been directly and bi-direc- tionally sequenced by using the corresponding primers for each amplified sequence. Since direct sequencing of G3pdh showed fragments of extra peaked sequencing data, we proceeded with cloning with InsTAclone PCR Cloning Kit (Thermo Scientific®) of the G3pdh amplifi- cation products. Several colonies for each cloned sam- ple were amplified using T7 and SP6 primers whose sites are located at the boundaries of the cloning region. PCR products were purified using the QIAquick PCR Purification Kit (Quiagen) and sent to the University of 29Possible hybrid speciation for two Malagasy species of Piper L. (Piperaceae) Florence internal sequencing service CIBIACI (www. cibiaci.unifi.it). Manual correction and assembly of the sequences was performed using the software Mul- talin (Corpet 1988) and MEGA7 (Kumar et al. 2016). Unexpectedly, two DNA sequences were obtained, after removing the cloning vector fragments, showing a dif- ferent size: 965bp and 1058bp which were named “short” and “long” sequences respectively (Figure 1). At a first sight only the long sequences of G3pdh have been considered as the right ones because, as Figure 1. Alignment of the long and short G3pdh sequences isolated from P .malgassicum and P. tsarasotrae using BioEdit software (Hall 1999). Shaded fragments represent the primers used for amplification. 30 Enrico Palchetti et al. observed by Smith et al. (2008), no paralogs have been detected in a great deal of other Piper species and there- fore other results have been discarded as PCR artifacts. In the present work a second thorough revision of the sequencing output has been carried on and showed the presence of overlapping peaks in all the samples and an additional round of analysis of the colonies con- firmed the presence of the short sequences. To rule out any doubt, two additional specimen for each species has been collected and submitted to amplification and cloning in order to confirm the presence of the short sequence. As all the samples showed the same pattern, we decide to use also this “short” sequence to study the phylogeny of these two piper species, by comparing with all the accession present online. The sequences used during our investigation are available in GenBank®: Piper tsarasotrae G3pdh long sequence (MH234634), G3pdh short sequence (MT793801), trnL (MH234638), ndhF (MH234636) and Piper malgassicum: G3pdh long sequence (MH234633), G3pdh short sequence (MT793800), trnL (MH234637), ndhF (MH234635). 2.1. Phylogenetic analysis The DNA sequences were aligned with CLUSTALX 2.0 and checked by eye for manual adjustment. The plas- tidial and the nuclear sequences were aligned separately to produce matrices that were later combined with the software combinex2_0.py (Python version 2.6.4; Biopy- thon 1.57), by A. Papini, released under GPL license and available at w w w.unifi.it/caryologia/PapiniPrograms. html as implemented in Bandara et al. (2013) and in Simeone et al. (2016). The phylogenetic analysis was executed on both cpDNA (ndhF and trnL) and nuclear sequences (G3pdh). Maximum parsimony analysis was performed with PAUP* 4.0b1 (Swofford 1998, 2001). The genbank sequences of P. humistratum Görts & K. U. were used as outgroups both in the nuclear and the plastid genes matrix, following the previous phylogenetic analysis by Smith et al. (2008). This sequences used as outgroup resulted belonging to the sister clade with respect to the clade containing the African species and the other relat- ed clades in Smith et al. (2008). References of the other species used in the analysis are summarized (with Gen- Bank® codes) in Table 1 in Smith et al. (2008). All char- acters had equal weight and unordered state transitions. Gaps were coded with the “simple indel coding” model (Simmons and Ochoterena 2000), with the software Gap- coder (Young and Healy 2003) and added to the final matrix after the DNA sequences as in Papini et al. (2004). The evolutionary model implemented in Mrbayes for treating gaps was the same as that proposed by Lewis et al. (2001) for treating morphological data, the Mk mod- el, justified as simple absence/presence of the character without a priori assignment of different weights. We used MrMODELTEST 2.0 (Nylander 2004) to choose the best evolutionary model of DNA sequences on the basis of the Akaike information criterion (Akai- ke 1974). The best model was used as settings with MrBayes 3.2.7 (Ronquist et al. 2012) for Bayesian Infer- ence. A maximum likelihood (ML) phylogenetic analy- sis was carried out with RaxML (Stamatakis 2014) and the resulting trees were edited with Figtree (Rambaut and Drummond 2010). We mapped the support on the tree branches with the results of the Bayesian phyloge- netic analysis after removing the first trees with low like- lihood values as “burn-in”, as in Papini et al. (Papini et al. 2007; Papini et al. 2011). The remaining trees were used to produce a 50% majority-rule consensus tree in which the percentage indicated on branches was used as a measure of the Bayesian posterior probability. 2.2. Karyological analysis Chromosomes images were obtained from somatic mitoses recorded from root tips of only one plant liv- ing in a pot. The procedure was the same as in Mosti et al. (2011) and Mousavi et al. (2013), with a pretreatment in 8-hydroxyquinoline and fixation in Carnoy. Then the material was hydrolyzed in HCl and then stained with Lacto-propionic-orceine. We observed metaphase plates of meristematic cells, with the technique of fresh squashes of root tips. Chro- mosome counts were made during direct observations with the microscope, and later recounted on enlarged digital images. Images were recorded with a microscope Leica DM RB Fluo. 3. RESULTS Amplification of two plastid fragments named ndhF and trnL intron was carried on and the amplicons cor- rectly sequenced producing reads of 1860 bp and 920 bp, respectively. Cloning of the amplicon of the nuclear gene G3pdh of P. malgassicum and P. tsarasotrae allowed to isolate two haplotypes, which were named “long” (1060bp for P. malgassicum and 1045bp for P. tsara- sotrae) and “short” 965 bp (for both species) after their size. We used a total of 71 sequences, considering sep- arately the short and long sequences of P. malgassicum and P. tsarasotrae for G3pdh and the plastid sequences 31Possible hybrid speciation for two Malagasy species of Piper L. (Piperaceae) matrix. The total alignment of the G3pdh region was 1127 nucleotides long including gaps. The final parts of the sequences were very variable and hence the align- ment was ambiguous. For this reason, we excluded the characters from position 957 to 1127. The rest of the alignment was used for indels (gap) coding (with the 0.02 26hostmannianum_frenchguiana 70pingbienense_china 71nudifolium_costarica 37arboreum_honduras 54capense_kenya 66betle_cultivated 34umbellatum_tanzania 1humistratum_frenchguiana 67betle_tanzania 27auritum_costarica 3austrocaledonicum_newcaledonia 57thomsonii_china 23concepcionis_costarica 40avellanum_frenchguiana 36arboreum_cultivated_ghana 12malgassicum short 68wallichii_china 39obliquum_nicaragua 18aduncum_dominicanrepublic 22guazacapanense_mexico 41costatum_cultivated 25colonense_nicaragua 42augustum_frenchguiana 13tsarasotrae short 16aduncum_nicaragua 17hispidum_mexico 64_vietnam 24pseudofuligineum_honduras 51santum_mexico2 60chaudocanum_china 59caninum_cultivated_australia 52sanctum_nicaragua 50sanctum_mexico3 8guineense_cameroon2 32umbellatum_mexico 7guineense_cameroon1 46puberulum_cultivated 19_tanzania 15rothianum_australia 21yucatanense_mexico 45guahamense_cultivated 62submultinerve_china 44aequale_honduras 63nigrum_cultivated 38imperiale_costarica 11tsarasotrae long 58subpenninerve_malaysia 56semiimmersum_china 65sarmentosum_cultivated 43urophyllum_costarica 61flaviflorum_china 48methysticum_hawaii2 31umbellatum_honduras 6guineense_uganda1 47methysticum_hawaii1 30peltatum_frenchguiana 35umbellatum_cameroon 9guineense_kenya 28auritum_mexico 4borbonense_cultivated_reunion 29peltatum_costarica 69hancei_china 55porphyrophyllum_malaysia 49sanctum_mexico1 10guineense_uganda2 53capense_cameroon 33umbellatum_kenya 20amalago_honduras 14muricatum_malaysia 100/100 89/100 99/100 85/100 100/100 100/10098/100 88/100 100/100 45/100 2malgassicum long Figure 2. Maximum likelihood tree produced by RAXML with nuclear sequences. The supports above or below the branches are, respec- tively, the bootstrap resampling support with maximum likelihood criterion produced by RAXML, and the bayesian support calculated including the information derived from indels. In case the bayesian support is lower than 50, it is not indicated on the figure. 32 Enrico Palchetti et al. 0.005 27aduncum_dominicanrepublic 1humistratum_frenchguiana 40imperiale_costarica 21guineense_kenya 71 pingbienense_china 8submultinerve_china 52costatum_cultivated 49umbellatum_mexico 14betle_tanzania 65capense_cameroon 20guineense_uganda2 7 flaviflorum_china 61sanctum_mexico1 13betle_cultivated 44peltatum_costarica 69subpenninerve_malaysia 55urophyllum_costarica 16borbonense_cultivated_reunion 3new_nigrum 57guahamense_cultivated 62sanctum_nicaragua 48umbellatum_tanzania 2new_unguiculatum 9wallichii_china 60methysticum_hawaii2 38arboreum_cultivated_ghana 46umbellatum_cameroon 41obliquum_nicaragua 15sarmentosum_cultivated 19tsarasotrae 36yucatanense_mexico 28concepcionis_costarica 6hancei_china 68 thomsonii_china 42auritum_costarica 32colonense_nicaragua 10nigrum_cultivated 22guineense_uganda1 25muricatum_malaysia 53augustum_frenchguiana 50peltatum_frenchguiana 64santum_mexico2 67semiimmersum_china 11piper_vietnam 66capense_kenya 29aduncum_nicaragua 63sanctum_mexico3 56aequale_honduras 24guineense_cameroon2 43auritum_mexico 17malgassicum 23guineense_cameroon1 33pseudofuligineum_honduras 35amalago_honduras 18caninum_cultivated_australia 58puberulum_cultivated 39arboreum_honduras 47umbellatum_kenya 45umbellatum_honduras 30hispidum_mexico 54nudifolium_costarica 51avellanum_frenchguiana 34hostmannianum_frenchguiana 26rothianum_australia 31piper_tanzania 37guazacapanense_mexico 59methysticum_hawaii1 5chaudocanum_china 70porphyrophyllum_malaysia 12austrocaledonicum_newcaledonia 64/96 98/100 77/ 86/100 90/100 80/98 96/85 /100 Figure 3. Maximum likelihood tree produced by RAXML with chloroplast sequences. The support indexes indicated on the tree are the same as in Figure 2 (maximum likelihood bootstrap and bayesian support). 33Possible hybrid speciation for two Malagasy species of Piper L. (Piperaceae) software gapcoder), resulting in further 99 characters that were inserted after the nucleotide sequences. The plastid genes ndhF and trnL were inserted one after the other in the sequence, producing an aligned matrix of 2016 characters. The coding of indels resulted in further 115 characters. RAxML applied on the nuclear G3pdh matrix (indels coding excluded) produced a maximum likelihood tree with bootstrap support obtained with 1000 replicates (Figure 2). The support on branches corresponds to maximum likelihood bootstrap support (left) and Bayesian support with gaps (on the right). The same method was using for the plastid matrix (Figure 3). Comparing the two maximum likelihood trees, the one based on nuclear DNA data (G3pdh sequences) and that obtained with plastid markers, we could observe that in the first case P. malgassicum, clustered together and as sister group of P. borbonense (Figure 2), another species from an island, La Reunion, which lies relatively close to Madagascar. This relationship is corroborated by 100% maximum likelihood bootstrap (MLS) and bayes- ian (BS) support. The other Malagasy species, P. tsara- sotrae, typical of arid forest, was more strictly related to the entries of the African species P. guineense, with 100% MLS and 100% BS. All these species formed a well characterized clade with 89% MLS and 100% BS and their closest species appeared to be Asian species P. cani- num, (Figure 2). The BS without considering gaps coding gave the same support in this clade. The “short” sequences of G3pdh of both P. tsara- sotrae and P. malgassicum clustered together within a group of Asian species, mainly originating from Malay- sia and Australia with 98% MLS and 100% BS (Figure 2). The (phylogenetic) story told by the data obtained from chloroplast genome sequences was quite differ- ent: the Malagasyan species P. tsarasotrae and P. mal- gassicum clustered together with the phytogeopraphi- cally close P. borbonense with 90% MLS and 100% BS, while the 5 accessions of the African P. guineense were in a more external condition with respect to the former group and separated in two groups, one from Cameroon (NW Africa) and one from Uganda/Kenya (Central-East Africa). All these species together formed a monophy- letic group with 64% MLS and 96% BS (95% bayesian support in the analysis without gaps). Also in this case P. caninum, together with P. rothianum, was the outgroup to the African + Malagasy species (Figure 3) with 80% MLS and 98% BS (99% without gaps). Adding indels data to the matrix did not appear to increase the support value of nodes in the plastidial genes tree. The counted chromosome numbers varied from 2n=46±2 in P. malgassicum (Figure 4A) to 2n=36± 2 in P. tsarasotrae (Figure 4B). The uncertainty in the counts, that should be taken only as preliminary result, derived from the small size of the chromosomes (many of them less than 1 μm of length), the low amount of metaphases in the root tips of the plants cultivated in Florence and the apparently small size of the mitotic spindle, leading A B Figure 4. Chromosomes. A) P. malgassicum number of chromosomes: 2n = about 46. Bar = 5 μm; B) P. tsarasotrae: 2n = about 38. Bar = 5 μm. 34 Enrico Palchetti et al. to partial overlapping of many of the small chromo- somes. Th e two currently known areal of the two species (two new localities discovered here) is shown in Figure 5. 4. DISCUSSION Th e fact that the phylogenetic history based on the chloroplast markers told a diff erent tale with respect to the tree produced with nuclear markers may be explained with a possible ancient hybridization/introgression event with pollen coming from an ancestor of the African P. guineense and reaching the ancestor of P. tsarasotrae, that would hence share some part of the nuclear genome with the African species. Th e only species of Piperaceae analyzed under the point of view of the type of plastid inheritance was a species of Peperomia, which resulted to have only maternal plastidial inheritance (Corriveau and Coleman 1988). Th e presence of the “short” G3pdh nuclear sequences may be related to a still more ancient hybridization event involving the ancestor of the Mala- gasy species and some ancestor of Asian origin. Also in this case probably, with Asian pollen entering in contact with the ancestoŕ s stigma of the Malagasyan species. As a matter of fact the closest relatives to the African spe- cies sensu lato (including the Malagasy and the Reun- ion species) are Asian, with the closest species (among those here sampled) apparently from Malaysia (Figure 3 and Figure 4). Apparently interspecifi c hybrids can be obtained in genus Piper also experimentally (Vanaja et al. 2008), while the hybrid origin of several Andean species was already proposed by Quijano-Abril et al. (2006). Th e presence of paralogs of G3pdh in angiosperms may represent a problem in several phylogenetic analy- sis (Hurteau and Spivack 2002); Liu et al. 2009; Sun et al 2012). However, here most of the indels were found in the introns of the gene and hence we are not able to assess the functionality of the short sequences. The preliminary results about the chromosome numbers scored about 2n=46+-2 in P. malgassicum and 2n=36+-2 in P. tsarasotrae. Th e uncertainty in the counts was due to the small dimensions of the chromo- somes that were observed in most of the species of the genus, together with stickiness (Samuel 1987; Samuel and Morawetz 1989), the low amount of metaphases in the root tips of the plants cultivated in vitro and the apparently small size of the mitotic fuse, leading to par- tial overlapping of many of the small chromosomes. Th e mitotic spindle can reach dimensions up to 60 μm (Wühr et al. 2008; Petry 2016), while in P. malgassicum and P. tsarasotrae it was about 15-20 μm (see Figure 4). Th e chromosome numbers in genus Piper are very variable, ranging from 2n=26 to 2n=104, with some species apparently able to possess several possible chro- mosome numbers (Samuel 1987). Most new world spe- cies show a karyotype of 2n=26 and x=13 (Samuel and Morawetz 1989), with some exceptions having 2n=28 chromosomes (Maugini 1953). In Asia tetraploids 2n=52 would prevail (Samuel 1987). No data was available for African and Malagasy species up to the here present- ed results. However, the clear diff erence in karyotype between P. tsarasotrae and P. malgassicum, two species otherwise strictly phylogenetically related, may confi rm a possible hybridization/introgression event with a spe- cies with a diff erent chromosome number with respect to the ancestor of the Malagasyan species. As a matter of fact, also Nair et al. (1993) explained the observation of a triploid plant of P. nigrum (2n=78) as the result of a natural crossing between 2n=52 and 2n=104 plants. Figure 5. Geographical localization of the sampling area for P. tsar- asotrae (yellow) and P. malgassicum (white). Area of the sampling campaign of 2018 (a and b) and 2019 (A and B) for P. tsaratsotrae (a/A) and P. malgassicum (b/B) respectively. 35Possible hybrid speciation for two Malagasy species of Piper L. (Piperaceae) The progeny showed a range of variation from 2n=52 to 2n=104 and production of aneuploid viable pollen (Nair et al. 1993). Hybridization may influence diversity, including gene flow from one taxon to another (intro- gression) and the formation of new, stable hybrid taxa and, possibly, speciation (Mallet 2007; Vallejo‐Marín and Hiscock 2016). As preliminary guess, the two different chromosome numbers of the Malagasyan species may have arisen as a consequence of hybridization of a 2n=52 species with a 2n=26 (P. tsarasotrae) and with another 2n=52 species (P. malgassicum), respectively, with following aneuploid reductions. The two species have close areals (they are almost sympatric), even if they tend to occupy different habitats, more arid P. tsarasotrae (not lianous habit) and more humid (lianous habit) P. malgassicum. Such proxim- ity of two closely related species may be considered anoth- er possible indication of a relatively fast speciation event. A discordance between plastid and nuclear inherit- ance inferred through DNA sequencing has been often related to reticulate evolution and species of hybrid ori- gin (García et al. 2014; Stefanović et al. 2007; Aubriot et al. 2018), even if the karyological data may be a decisive evidence, it has been rarely used in relationship to the DNA sequence evidence as, for instance in Selvi et al. (2002). Here the hypothesis of an hybrid origin for the two investigated species may explain the presence of a double G3pdh sequence in both of them. The relationship between the Malagasyan species and P. guineense with the asian species as members of Piper s. s. was already proposed by Jaramillo and Calle- jas (2004) and Jaramillo et al. (2008) as a result of a dis- persal event, and our results do not disagree with this position. Apparently, in Africa and Madagascar the con- ditions leading to the wide diversification observed in South-american Piper (Martines et al. 2015) are lacking or less capable of influencing the speciation process. 5. CONCLUSIONS The surprising discrepancy between the nuclear and the plastid phylogeny could be explained with an ancestral introgression event due probably to pollen contribution from an ancestor of the African mainland P. guineense towards the ancestor of P. tsarasotrae. The presence of possible paralogs of the nuclear gene G3pdh, clustering together with more distantly related Asian species lead to the hypothesis that a second more ancient hybridation/introgression event would have occurred between south Asian species and the ancestor of the Malagasyan species. The chromosome numbers observed in the Malagasyan species would confirm different evo- lutionary history. Further studies about the karyotypes of the Mala- gasy species, the african P. guineense and P. borbonense will be necessary together with the investigation of the possible presence of short paralog sequences of G3pdh in P. borbonense. ACKNOWLEDGMENTS the Authors would like to acknowledge Dr. Nicola Gandolfi from the NGO TsiryParma, Ambositra, Mada- gascar, for the support given in the germplasm collection in field. REFERENCES Akaike H (1974) A new look at the statistical model identification. IEEE Trans Autom Control 19: 716– 723. https://doi.org/10.1109/TAC.1974.1100705 Aubriot X, Knapp S, Syfert MM, Poczai P and Buerki S (2018) Shedding new light on the origin and spread of the brinjal eggplant (Solanum melongena L.) and its wild relatives. Am J Bot 105: 1175–1187. https:// doi.org/10.1002/ajb2.1133 Bandara N, Papini A, Mosti S, Brown T, Smith L (2013) A phylogeny of Onobrychis and its relationships with allied genera of Hedysareae. Turk Journ of Bot. 37(6): 981-992. https://doi.org/10.3906/bot-1210-32 Bressan E A, Rossi ML, Lee TSG, Figueira A (2014) Extraction of high-quality DNA from ethanol-pre- served tropical plant tissues.” BMC research notes 7: 268. https://doi.org/10.1186/1756-0500-7-268 Burger W (1972) Evolutionary trends in the Central American Species of Piper (Piperaceae). Brittonia 24: 356-362. https://doi.org/10.2307/2805498 Corpet F (1988) Multiple sequence alignment with hier- archical clustering. Nucleic Acids Res 16:10881– 10890. https://doi.org/10.1093/nar/16.22.10881 Corriveau JL, Coleman AW (1988) Rapid screening method to detect potential biparental inheritance of plastid DNA and results for over 200 angio- sperm species. Am J Bot 75: 1443–1458. https://doi. org/10.2307/2444695 De Candolle C (1869) Piperaceae. In: De Candolle C, edi- tor. Prodromus Systematis Naturalis Regni Vegetabi- lis, Vol. 16. Part 1. Paris, France: Masson, pp. 235-471 (in Latin). De Candolle C (1923) Piperacearum clavis analytica. Candollea 1: 65-415 (in Latin). 36 Enrico Palchetti et al. García MA, Costea M, Kuzmina M, Stefanović S (2014) Phylogeny, character evolution, and biogeography of Cuscuta (dodders; Convolvulaceae) inferred from coding plastid and nuclear sequences. Am J Bot 101:670–690. doi:10.3732/ajb.1300449 Hall TA (1999) BioEdit: A User-Friendly Biological Sequence Alignment Editor and Analysis Program for Windows 95/98/NT. Nucleic Acids Symposium Series, 41: 95-98. https://doi.org/10.14601/Phyto- pathol_Mediterr-14998u1.29 Hurteau GJ, Spivack SD (2002) mRNA-specific reverse transcription-polymerase chain reaction from human tissue extracts. Anal Biochem 307:304-315. https:// doi.org/10.1016/S0003-2697(02)00058-1 Isnard S, Prosperi J, Wanke S, Wagner ST, Samain M-S, Trueba S, Frenzke L, Neinhuis C, Rowe NP (2012) Growth form evolution in Piperales and its relevance for understanding Angiosperm diversification: an integrative approach combining plant architecture, anatomy, and biomechanics. Int J Plant Sci 173: 610- 639. https://doi.org/10.1086/665821 Jaramillo MA and Callejas R (2004) Current perspectives on the classification and phylogenetics of the genus Piper L. Pp. 179–198. in Piper: A model genus for studies of chemistry, ecology, and evolution, eds. L. A. Dyer and A. N. Palmer. Boston: Kluwer Academic. Jaramillo MA, Callejas R, Davidson C, Smith JF, Ste- vens AC, and Tepe EJ (2008) A Phylogeny of the Tropical Genus Piper Using ITS and the Chloroplast Intron psbJ–petA. Systematic Botany 33(4): 647–60. doi:10.1600/036364408786500244. Jaramillo MA, Manos PS (2001) Phylogeny and pat- terns of floral diversity in the genus Piper (Piperaceae). Am J Bot 88:706–716. https://doi. org/10.2307/2657072 Kumar S, Stecher G, Tamura K (2016) MEGA7: Molecu- lar Evolutionary Genetics Analysis Version 7.0 for Bigger Datasets. Mol Biol Evol 33: 1870–1874. htt- ps://doi.org/10.1093/molbev/msw054 Larkin MA, Blackshields G, Brown NP, Chenna R, McGettigan PA, McWilliam H, Valentin F, Wallace IM, Wilm A, Lopez R, Thompson JD, Gibson TJ, Higgins DG (2007) Clustal W and Clustal X ver- sion 2.0. Bioinformatics 23: 2947-2948. https://doi. org/10.1093/bioinformatics/btm404 Lewis PO (2001) A likelihood approach to estimat- ing phylogeny from discrete morphological char- acter data. Syst Biol 50: 913–925. https://doi. org/10.1080/106351501753462876 Liu Y-J, Zheng D, Balasubramanian S, Carriero N, Khurana E, Robilotto R, Gerstein MB (2009) Com- prehensive analysis of the pseudogenes of glycolytic enzymes in vertebrates: the anomalously high num- ber of GAPDH pseudogenes highlights a recent burst of retrotranspositional activity. BMC Genomics 10: 480. https://doi.org/10.1186/1471-2164-10-480 Loconte H, Stevenson DW (1991) Cladistics of the Magnoliidae. Cladistics 7:267–296. https://doi. org/10.1111/j.1096-0031.1991.tb00038.x Mallet J (2007) Hybrid speciation. Nature 446: 279– 283. Martines C, Carvalho MR, Marinan S, Jaramillo CA (2015) A Late Cretaceous Piper (Piperaceae) from Colombia and diversification patterns for the genus. Am J Bot 102(2): 273-289. https://doi.org/10.3732/ ajb.1400427 Maugini E. (1953) Ricerche Cito-Embriologiche su Piper Medium Jacq. var. ceanothifolium (HBK) Trel. et Yun: (con 14 figure nel testo). Caryologia 5(3): 282-287. Mosti S., Fiorini G. and Papini A. (2011) Karyological investigations on several species of genus Rebutia Sect. Digitorebutia (Cactacee). Caryologia 64(3):350- 359 https://doi.org/10.1080/00087114.2011.10589802 Mousavi S H, Hassandokht M R, Choukan R, Sepahvand N, Khosrowchahli M, Papini A. (2013) Cytological study of chromosome and genome composition of Iranian lettuce (Lactuca sativa L.) accessions. Caryo- logia 66(1): 41-49. Digitorebutia (Cactaceae). Caryo- logia 64(3): 302-308. https://doi.org/10.1080/0008711 4.2013.780440 Murray MG and Pitas JW (1996) “Plant DNA from alcohol-preserved samples.” Plant molecular biol- ogy reporter 14: 261-265. https://doi.org/10.1007/ BF02671661 Nair RR, Sasikumar B, Ravindran PN (1993) Polyploidy in a cultivar of black pepper (Piper nigrum L.) and its open pollinated progenies. Cytologia 58: 27–31. Nylander JAA, (2004) Mr Modeltest, version 1.0b. Department of Systematic Zoology, EBC, Uppsala University, Uppsala, Sweden. Palchetti E, Biricolti S, Gori M., Rota Nodari G, Gan- dolfi N, Papini A (2018) Two new Malagasy species of genus Piper L. (Piperaceae): Piper malgassicum and Piper tsarasotrae and their phylogenetic position. Turk J of Bot 42(5): 622-610. doi:10.3906/bot-1712-2 Papini A, Banci F, Nardi E (2007) Molecular evidence of polyphyletism in the plant genus Carum L. (Api- aceae). Genet Mol Biol 30: 475–482. https://doi. org/10.1590/S1415-47572007000300029 Papini A, Simeone MC, Bellarosa R, Spada F, Schirone B (2011) Quercus macranthera Fisch. & Mey. Ex Hohen. and Quercus iberica M. Bieb.: taxonomic definition and systematic relationships with Euro- pean oaks inferred from nuclear Internal Transcribed Spacer (ITS) data. Plant Biosyst 145: 37-49. https:// 37Possible hybrid speciation for two Malagasy species of Piper L. (Piperaceae) doi.org/10.1080/11263504.2010.502684 Papini A, Trippanera GB, Maggini F, Filigheddu R, Bion- di E (2004) New insights in Salicornia L. and allied genera (Chenopodiaceae) inferred from nrDNA sequence data. Plant Biosystems 138(3): 215-223 htt- ps://doi.org/10.1080/11263500400006977 Petry S (2016) Mechanisms of Mitotic Spindle Assem- bly. Annu Rev Biochem 85:659–683. https://doi. org/10.1146/annurev-biochem-060815-014528 Quijano-Abril MA, Callejas-Posada R, Miranda-Esquivel DR (2006) Areas of endemism and distribution pat- terns for Neotropical Piper species (Piperaceae). J Biogeogr 33:1266–1278. https://doi.org/10.1111/ j.1365-2699.2006.01501.x Rambaut A, Drummond A (2010) FigTree v1.3.1. http:// tree.bio.ed.ac.uk/software/figtree. Institute of Evolu- tionary Biology, University of Edinburgh, Edinburgh, United Kingdom. Ronquist F, Teslenko M, Van Der Mark P, Ayaccesseres DL, Darling A, Höhna S, Larget B, Liu L, Suchard MA, Huelsenbeck JP (2012) Mrbayes 3.2: efficient bayesian phylogenetic inference and model choice across a large model space. Syst Biol 61: 539–42. htt- ps://doi.org/10.1093/sysbio/sys029 Samuel R (1987) Chromosome Numbers in Piper. Kew Bull 42: 465. https://doi.org/10.2307/4109710 Samuel R, Morawetz W (1989) Chromosomal evolution within Piperaceae. Pl. Syst. Evol. 166: 105–117. Selvi F, Papini A, Bigazzi M (2002) Systematics of Nonea (Boraginaceae-Boragineae): new insights from phe- netic and cladistic analyses. Taxon 51: 719–730. htt- ps://doi.org/10.2307/1555025 Simeone MC, Grimm GW, Papini A, Vessella F, Cardoni S, Tordoni E, Piredda R, Franc A, Denk T (2016) Plastome data reveal multiple geographic origins of Quercus Group Ilex. PeerJ 4:e1897 https://doi. org/10.7717/peerj.1897 Simmons MP, Ochoterena H (2000) Gaps as Characters in Sequence-Based Phylogenetic Analyses. Syst Biol 49: 369–381. https://doi.org/10.1093/sysbio/49.2.369 Smith JF, Stevens AC, Tepe EJ, Davidson C (2008) Plac- ing the origin of two species-rich genera in the late cretaceous with later species divergence in the ter- tiary: a phylogenetic, biogeographic and molecular dating analysis of Piper and Peperomia (Piperaceae). Plant Syst Evol 275:9–30. https://doi.org/10.1007/ s00606-008-0056-5 Stamatakis A (2014) RAxML version 8: a tool for phy- logenetic analysis and post-analysis of large phy- logenies. Bioinformatics 30: 1312–1313. https://doi. org/10.1093/bioinformatics/btu033 Stefanović S, Kuzmina M, and Costea M (2007) Delimi- tation of major lineages within Cuscuta subgenus Grammica (dodders; Convolvulaceae) using plastid and nuclear DNA sequences. Am J Bot 94: 568– 589. https://doi.org/10.3732/ajb.94.4.568 Strand AE, Leebens‐Mack J, Milligan BG (1997) Nuclear DNA‐based markers for plant evolutionary biology. Mol Ecol 6: 113–118. https://doi.org/10.1046/j.1365- 294X.1997.00153.x Sun Y, Li Y, Luo D, Liao DJ (2012) Pseudogenes as weak- nesses of ACTB (Actb) and GAPDH (Gapdh) used as reference genes in reverse transcription and polymer- ase chain reaction. PLoS ONE 7:e41659. https://doi. org/10.1371/journal.pone.0041659 Suwanphakdee C, Simpson DA, Hodkinson TR, Chan- taranothai P (2016) Taxonomic notes on the genus Piper (Piperaceae). Nord J Bot 34:605–618. https:// doi.org/10.1111/njb.01114 Swofford DL (1998) PAUP* 4.1. Phylogenetic Analysis Using Parsimony. Test version. Sunderland (MA): Sinauer Associates. Swofford DL (2001) Laboratory of Molecular Systematics Smithsonian Institution. 143 Taberlet P, Gielly L, Pautou G, Bouvet J (1991) Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Mol Biol 17: 1105–1109. https://doi.org/10.1007/BF00037152 Tebbs MC (1993) Piperaceae. In: Kubitzki K, Rohwer JG, Bittrich V, editors. Flowering Plants Dicotyledons. The Families and Genera of Vascular Plants, Vol. 2. Berlin, Germany: Springer, pp. 516-520. Ulloa Ulloa C, Acevedo-Rodríguez P, Beck S, Belgrano MJ, Bernal R, Berry PE, Brako L, Celis M, Davidse G, Forzza RC, Gradstein SR, Hokche O, Leon B, Leon-Yanez S, Magill RE, Neill DA, Nee M, Raven PH, Stimmel H, Strong MT, Villaseñor JL, Zarucchi JL, Zuloaga FO, Jørgensen PM (2017) An integrated assessment of the vascular plant species of the Amer- icas. Science 358:1614–1617. https://doi.org/10.1126/ science.aao0398 Vallejo‐Marín M, Hiscock SJ (2016) Hybridization and hybrid speciation under global change. New Phytolo- gist 211(4): 1170-1187. Vanaja T, Neema VP, Mammootty KP, Rajeshkumar R (2008) Development of a promising interspecific hybrid in black pepper (Piper nigrum L.) for Phy- tophthora foot rot resistance. Euphytica 161:437–445. https://doi.org/10.1007/s10681-007-9602-4 Weil M, Shum Cheong Sing A, Méot JM, Boulanger R, Bohuon P (2017) Impact of blanching, sweating and drying operations on pungency, aroma and color of Piper borbonense. Food Chem 219:274–281. https:// doi.org/10.1016/j.foodchem.2016.09.144 38 Enrico Palchetti et al. Wühr M, Chen Y, Dumont S, Groen AC, Needleman DJ, Salic A, Mitchison TJ (2008) Evidence for an Upper Limit to Mitotic Spindle Length. Curr Biol 18:1256– 1261. https://doi.org/10.1016/j.cub.2008.07.092 Young ND, Healy J (2003) GapCoder automates the use of indel characters in phylogenetic analysis. BMC Bioinformatics 4: 6 https://doi.org/10.1186/1471- 2105-4-6